SYMPTOM-BASED DISEASE PREDICTION USING MACHINE LEARNING

There are now more opportunities to increase diagnostic accessibility and accuracy thanks to the application of machine learning (ML) in healthcare, especially in environments with limited resources. The Random Forest Classifier (RFC) and Multi-Layer Perceptron (MLP) models emphasize this study's strong framework for symptom-based disease prediction utilizing machine learning methods. Our approach emphasizes the significance of data preparation, feature engineering, and model evaluation while addressing important issues, including missing data, symptom overlap, and ethical implications using Kaggle datasets. According to our findings, the RFC model performs better than the MLP classifier, with 99% accuracy. We also created an interactive platform for disease prediction, data addition, and model retraining using a web application built using Streamlit. Especially in poverty-stricken areas, this approach provides a scalable and dependable tool for early disease diagnosis, lowering diagnostic mistakes and enhancing healthcare accessibility.